This research is directed at providing an ultrastructural description of functional circuits in the cat's lateral geniculate nucleus, particularly in the A-laminae. This will enable us to better understand the means by which geniculate circuitry achieves its main function, which is to control the flow of information from retina to cortex. That is, the vast majority (80-90%) of synaptic input onto geniculate relay cells that is nonretinal serves to gate or control the gain of retinogeniculate transmission. There are many different ways in which this can be achieved, and it seems likely that these represent the neural substrates for many (but not all) forms of visual attention. Our main technique will continue to be electron microscopic reconstruction of single cells and/or axons labeled either with intracellular iontophoresis of horseradish peroxidase or with extracellular application of phaseolus vulgaris leucoagglutinin. We wish to obtain a complete description of the patterns and sources of synaptic inputs to relay X and Y cells plus interneurons. We shall focus on reliable differences seen among cell types, which also requires a description of the variability seen among cell classes. We ultimately want to understand the relationship between retinal and nonretinal inputs formed onto relay cells in order to better understand how the nonretinal inputs might affect retinogeniculate transmission. Nonretinal sources of input that will be investigated include: local collaterals of relay cell axons; interneurons; the perigeniculate nucleus; the visual cortex; and the brainstem reticular formation, including the parabrachial region, the locus coeruleus, and the raphe nucleus. Immunocytochemistry will be used in an attempt to reveal the putative transmitters used by certain synaptic inputs, such as: gamma-aminobutyric acid by interneurons and perigeniculate cells; acetylcholine, noradrenalin, and serotonin by brainstem axons; and glutamate (or a related substance) by retinal and cortical axons. We also hope to develop immunocytochemical techniques to identify the postsynaptic receptors, such as GABAA vs. GABAB, NMDA vs. non- NMDA, and muscarinic vs. nicotinic. Finally, we shall develop an in vitro preparation to label elements not easily labeled in vivo. This includes the GABAergic interneurons and perigeniculate cells plus various neuron pairs that might be synaptically connected.